<?xml version="1.0" encoding="utf-8" standalone="no"?>
<!DOCTYPE article SYSTEM "http://www.nonlin-processes-geophys.net/inc/npg/copernicus.dtd">
<article language="en">
	<journal>
		<journal_title>Nonlinear Processes  in Geophysics</journal_title>
		<journal_url>www.nonlin-processes-geophys.net</journal_url>
		<issn>1023-5809</issn>
		<eissn>1607-7946</eissn>
		<volume_number>10</volume_number>
		<issue_number>6</issue_number>
		<publication_year>2003</publication_year>
	</journal>
	<doi>10.5194/npg-10-525-2003</doi>
	<article_url>http://www.nonlin-processes-geophys.net/10/525/2003/</article_url>
	<abstract_html>http://www.nonlin-processes-geophys.net/10/525/2003/npg-10-525-2003.html</abstract_html>
	<fulltext_pdf>http://www.nonlin-processes-geophys.net/10/525/2003/npg-10-525-2003.pdf</fulltext_pdf>
	<start_page>525</start_page>
	<end_page>529</end_page>
	<publication_date>0000-00-00</publication_date>
	<article_title content_type="html">A deterministic width function model</article_title>
	<authors>
		<author numeration="1" affiliations="1">
			<name>C. E. Puente</name>
		</author>
		<author numeration="2" affiliations="">
			<name>B. Sivakumar</name>
		</author>
	</authors>
	<affiliations>
		<affiliation numeration="1" content_type="html">Department of Land, Air &amp; Water Resources, University of California, Davis, USA</affiliation>
	</affiliations>
	<abstract content_type="html">Use of a
      deterministic fractal-multifractal (FM) geometric method to model width
      functions of natural river networks, as derived distributions of simple
      multifractal measures via fractal interpolating functions, is reported. It
      is first demonstrated that the FM procedure may be used to simulate
      natural width functions, preserving their most relevant features like
      their overall shape and texture and their observed power-law scaling on
      their power spectra. It is then shown, via two natural river networks (Racoon
      and Brushy creeks in the United States), that the FM approach may also be
      used to closely approximate existing width functions.</abstract>
	<references>
	</references>
</article>

